Neuroscience Letters, 144 (1992) 165-168
165
© 1992 Elsevier Scientific Publishers Ireland Ltd. All rights reserved 0304-3940/92/$ 05.00 NSL 08946
Selective potentiation of GABA-mediated C1- current by lanthanum ion in subtypes of cloned GABAA receptors M a r v i n S. I m a, B e v e r l y J. H a m i l t o n b, D o n a l d B. C a r t e r b a n d W h a Bin I m b aKalamazoo College and bCNS Diseases Research, The Upjohn Company, Kalamazoo, M149001 (USA)
(Received 13 May 1992; Revised version received 15 June 1992; Accepted 17 June 1992) Key words." ClonedGABAAreceptor; Subunit composition; Lanthanum; Selectivepotentiation; GABA-induced C1- current
The effect of lanthanum ion (La3+)on 7-aminobutyric acid (GABA)-mediatedCI currents was examined in the a~82 or cqfl#2 subtype of GABAA receptors expressedin a human kidney cell line (A293), using a whole-cellconfiguration of patch-clamptechniques. La3+dose-dependentlystimulated the CI- currents in the ~8#2 subtype with an ECs0of 21.3+3.5/~M with a maximal potentiation of 240_+16% as normalized to the GABA response at 5/IM. In the ~82 subtype, however, the ion marginally potentiated GABA response, a maximal stimulation being less than 70% with an ECs0for La3+near 200/~M. The stimulation of GABA response by La3÷in the ~8#2 subtype was due to a decrease in the half maximal concentration for GABA and was more pronounced at the negative membrane potentials. This selectivityof La3+toward the subtypes of GABAAreceptors contrasts to that of Zn2+which inhibits the currents in the aLfl2,but not in the a,fl#2 subtype (Neuron, 5: (1990) 781 788). It appears that these polyvalent cations are useful in understanding the molecular basis for the functional diversityand in characterizing the molecular organization of native GABAA receptors.
The 7-aminobutyric acidA (GABAA) receptor is a supermolecular receptor-C1- channel complex made of multimeric subunits [2, 7, 10, 16]. A large family of the receptor subunits are known to exist in mammalian brains through cloning of the c D N A s and in situ hybridization studies [5, 7, 12, 13, 17, 23, 25]. Various conbinations of the subunits were expressed in X e n o p u s laevis oocytes and embryonic h u m a n kidney cells [7, 12-14, 18, 21]. In particular, the cloned receptors made of ot~, f12 and 72 subunits in h u m a n embryonic kidney cells displayed pharmacological properties identical to Type I receptors in brain tissue [12], and electrophysiological characteristics ( G A B A sensitivity, desensitization and C1- conductances) similar to those on neurons [14, 18, 21]. Changes in the subunit composition produced cloned receptors of differential selectivity toward benzodiazepines and various non-benzodiazepine hypnotics [12-14, 18]. Recently, polyvalent cations such as Zn 2+ and La 3÷ have been reported to influence the GABA-mediated C1- currents in diverse neuronal preparations [1, 6, 9, 15, 22, 24]. In particular, Zn 2* has been shown to selectively inhibit the currents in the receptors made of 0~Bz,but not that of Correspondence:W.B. Im, CNS Research 7251-209-1, The Upjohn Company,301 Henrietta St., Kalamazoo, M149001, USA. Fax: (1)(616) 385-4525.
a~2~/2 subunits
[3, 20]. We initiated this study to see whether La 3+, which augments the G A B A response [15, 23] in contrast to the inhibitory action of Zn 2+ [1, 3, 6, 9, 19, 20, 22], displays the same selectivity toward the subtypes of GABAA receptors. Information on subunit selectivity of inorganic ions as well as organic drugs (i.e. benzodiazepines) is useful to understand molecular characteristics of G A B A receptors as well as to study regional and developmental diversity of neuronal GABAA receptors. The stable cell lines expressing 0h [5], ~2 [25] and ?'2 [17] subunits of rat G A B A A receptors were derived by calcium-mediated transfection of plasmids containing a~, f12 and 7/2 c D N A and a plasmid encoding G418 resistance into human kidney cells (A293) [4]. D N A manipulations were carried out with standard methods using the transfer vector pVL-941 [8]. A293 cells were grown in modified Eagle's minimum essential medium supplemented with 10% fetal calf serum under humidified air containing 5% CO2 [11]. After two weeks of selection in 1 mg/ml G418, resistant cells were assayed for the ability to synthesize all 3 G A B A A receptor m R N A s by Northern blotting. The whole cell configuration of the patch clamp technique was used to record the GABA-mediated CIcurrents in A293 cells expressing the cloned G A B A A receptor, cqfl, y 2 or a.fl2, following the experimental proce-
166
dures described elsewhere [3]. Briefly, patch pipettes made of borosilicate glass tubes were fire-polished and showed a resistance of 0.5 2 M.Q when filled with a solution containing (mM) CsC1 140. E G T A I I. MgCle 2, ATP 0.5 and HEPES 10, pH 7,3. The cell-bathing external solution contained (mM) NaCI 135, KC1 5, MgC12 1, CaCL 1.8 and HEPES 5, p H 7.2, (normal saline). G A B A at the concentration of 5/aM in the external solution with or without indicated drugs was applied through a U-tube placed within 100/am of the cells. The current was recorded with an Axopatch 1D amplifier and a CV-4 headstage (Axon Instrument Co.). Bh-1 bath headstage was used to compensate changes in bath potentials. The currents were recorded with a Gould Recorder 220. G A B A currents were measured at the holding potential of - 6 0 mV at room temperature (21 24°C). The dose response curves were analyzed using a logistic equation,
%13272 {LaS']laM 5 ~MGABA
5
I0
50
I00
f. 500 pA
alg2
4o0 8oo" [25opa
ILa~']~M 5 ~MGABA
250
!
I
10
100
200
~ '
0
Ic, = 1..... x [Substrate]"/EC50" + [Substrate]")
(1) 0
where Ic, is the amplitude of the peak C1 current at the given substrate concentration ([Substrate]), Im~xis a maximal response and n is a Hill coefficient. Data fitting was carried out using a computer program (Sigma Plot) of non-linear regression. Data are presented as the mean + S.E.M. from 3 5 separate measurements. Experimental variations in results from different cells were less than 15%. Fig. 1 shows the effect of La -*~ on GABA-mediated CI currents in the cells expressing the o~,flzY2 or 0t~2 subtype of GABAA receptors. In the oq]32T 2 subtype, La > dosedependently potentiated the G A B A (5 /aM) response. Analysis of the dose-response curve using Equation 1 provided the following binding parameters for La>: an EDs0 of 21.3_+3.5/aM, a Hill coefficient of 1.4_+0.4 and the maximal stimulation of 240+16% as normalized to the response by G A B A at 5/aM. In the ot~fle subtype, on the other hand, La > showed no noticeable stimulation of G A B A response until its concentration was raised above 100/aM: Fig. 1 shows the dose-response curve for La > in the a ~ 2 subtype obtained with G A B A at 1/aM, which was similar to that observed with G A B A at 5/aM. The maximal potentiation of the G A B A response by La ~+ reached around 70% at its concentration of 800/aM. This indicates that the 0~2 subtype is lacking a high affinity site for La > associated with the alf1272 subtype. This selectivity of La > is quite opposite to that of Zn 2+ which inhibited the G A B A response in the ~ 6 , with only a marginal action on the ~tfl2Y2 subtype [3, 20]. In this study, we further characterized the potentiation of G A B A currents by La > in the (~1fl272 subtype. Fig. 2 shows the degrees of potentiation by La > (50/aM) as a function of G A B A concentration. The reciprocal plots of
150
0 ~
100
0
50
*P,4
0 1000
[La3+] u Fig. 1. Differential potentiation of GABA-gatedcurrents by La> in the ~qfl2and ottf1272subtypes of GABAAreceptors. The upper panel shows the representative traces for GABA-induced C1 currents in the presence of various concentrations of La> as measured in A293 cells expressing the subtypes, using the whole cell configuration of the patch clamp techniques. The bottom panel shows dose-response curves for La~+potentiation in the presence of 5/aM GABA in ~L//27:(~3)and 1 /aM GABA in ottfl2 (e). The data were fitted to Equation 1and represent the mean and standard errors from 3 cells. [GABA] vs. It, (Fig. 2) show that the stimulatory effect of La 3+ at 50/aM progressively decreased as the concentration of G A B A increased. The analysis of the data with Equation 1 revealed that the major effect of La '~ (50/aM) was to reduce the half maximal concentration for G A B A from 19.3+1 to 12.6+1.1 /aM (n=3) without noticeable effects on the maximal amplitude of the CI- currents and the Hill coefficient (2.6_+0.4 vs. 2.4_+0.4). Although its primary effect was to affect G A B A affinity, La > is not likely to interact directly with G A B A site(s), because G A B A site(s) should be c o m m o n in the both subtypes of GABAA receptors and could not be a contributing factor
167 0.06 IA! s*
GAB^ 2 IIM -
50 IIM 0
250~ GABA
?
Control
• Las+
/
/
ov 0,04
10~M~
.60.,0.,0
o
.~ l
c~
U
GAB^~0.M ~f-
0,02
0.0
0.4
0.8
,-a
0.00 1.2
~ U
Fig. 2. Double-reciprocalplots of GABA-gated currents vs. GABA concentrationswith or without La 3÷. The left panel shows the representative traces for CI currents induced in the presenceof 2, 5, 10 and 20 /.tM GABA with or without 50/.tM La 3÷ as measured in A293 cells expressing%8272subunits. The right panel shows the double reciprocal plots of GABA-induced currents (I,:0 vs. GABA concentrations with (e) or without (o) 50/.tM La3+. for the subtype selectivity. La 3+ may exert its action through an allosteric interaction with a site not shared by the two subtypes. One of the most prominent allosteric site unique to the afl2Y2 subtype is benzodiazepine binding sites [12 14, 18], but it does not appear to be the La 3+ binding site, because excess Ro 15-1788, a classical benzodiazepine antagonist, was unable to block the potentiation of GABA currents by La 3+ (data not shown). Fig. 3 shows typical current-voltage if-V) relationships for the GABA (5/IM)-induced C1- currents with or without La 3÷ at 50 ~tM. La 3+ showed no effect on the reversal potential, 4_+3 mV (n=4), but its potentiation was considerably attenuated at the negative membrane potentials. As reported earlier [21], the afl2Y2 subtype showed an outward rectification in the I - V relationship. The ratio of chord conductances at +40/-40 mV was 2.9+0.4 and decreased to 1.6_+0.3 in the presence of La 3+ at 50/JM. This significant decrease in outward rectification in the presence of La 3- (P
165
© 1992 Elsevier Scientific Publishers Ireland Ltd. All rights reserved 0304-3940/92/$ 05.00 NSL 08946
Selective potentiation of GABA-mediated C1- current by lanthanum ion in subtypes of cloned GABAA receptors M a r v i n S. I m a, B e v e r l y J. H a m i l t o n b, D o n a l d B. C a r t e r b a n d W h a Bin I m b aKalamazoo College and bCNS Diseases Research, The Upjohn Company, Kalamazoo, M149001 (USA)
(Received 13 May 1992; Revised version received 15 June 1992; Accepted 17 June 1992) Key words." ClonedGABAAreceptor; Subunit composition; Lanthanum; Selectivepotentiation; GABA-induced C1- current
The effect of lanthanum ion (La3+)on 7-aminobutyric acid (GABA)-mediatedCI currents was examined in the a~82 or cqfl#2 subtype of GABAA receptors expressedin a human kidney cell line (A293), using a whole-cellconfiguration of patch-clamptechniques. La3+dose-dependentlystimulated the CI- currents in the ~8#2 subtype with an ECs0of 21.3+3.5/~M with a maximal potentiation of 240_+16% as normalized to the GABA response at 5/IM. In the ~82 subtype, however, the ion marginally potentiated GABA response, a maximal stimulation being less than 70% with an ECs0for La3+near 200/~M. The stimulation of GABA response by La3÷in the ~8#2 subtype was due to a decrease in the half maximal concentration for GABA and was more pronounced at the negative membrane potentials. This selectivityof La3+toward the subtypes of GABAAreceptors contrasts to that of Zn2+which inhibits the currents in the aLfl2,but not in the a,fl#2 subtype (Neuron, 5: (1990) 781 788). It appears that these polyvalent cations are useful in understanding the molecular basis for the functional diversityand in characterizing the molecular organization of native GABAA receptors.
The 7-aminobutyric acidA (GABAA) receptor is a supermolecular receptor-C1- channel complex made of multimeric subunits [2, 7, 10, 16]. A large family of the receptor subunits are known to exist in mammalian brains through cloning of the c D N A s and in situ hybridization studies [5, 7, 12, 13, 17, 23, 25]. Various conbinations of the subunits were expressed in X e n o p u s laevis oocytes and embryonic h u m a n kidney cells [7, 12-14, 18, 21]. In particular, the cloned receptors made of ot~, f12 and 72 subunits in h u m a n embryonic kidney cells displayed pharmacological properties identical to Type I receptors in brain tissue [12], and electrophysiological characteristics ( G A B A sensitivity, desensitization and C1- conductances) similar to those on neurons [14, 18, 21]. Changes in the subunit composition produced cloned receptors of differential selectivity toward benzodiazepines and various non-benzodiazepine hypnotics [12-14, 18]. Recently, polyvalent cations such as Zn 2+ and La 3÷ have been reported to influence the GABA-mediated C1- currents in diverse neuronal preparations [1, 6, 9, 15, 22, 24]. In particular, Zn 2* has been shown to selectively inhibit the currents in the receptors made of 0~Bz,but not that of Correspondence:W.B. Im, CNS Research 7251-209-1, The Upjohn Company,301 Henrietta St., Kalamazoo, M149001, USA. Fax: (1)(616) 385-4525.
a~2~/2 subunits
[3, 20]. We initiated this study to see whether La 3+, which augments the G A B A response [15, 23] in contrast to the inhibitory action of Zn 2+ [1, 3, 6, 9, 19, 20, 22], displays the same selectivity toward the subtypes of GABAA receptors. Information on subunit selectivity of inorganic ions as well as organic drugs (i.e. benzodiazepines) is useful to understand molecular characteristics of G A B A receptors as well as to study regional and developmental diversity of neuronal GABAA receptors. The stable cell lines expressing 0h [5], ~2 [25] and ?'2 [17] subunits of rat G A B A A receptors were derived by calcium-mediated transfection of plasmids containing a~, f12 and 7/2 c D N A and a plasmid encoding G418 resistance into human kidney cells (A293) [4]. D N A manipulations were carried out with standard methods using the transfer vector pVL-941 [8]. A293 cells were grown in modified Eagle's minimum essential medium supplemented with 10% fetal calf serum under humidified air containing 5% CO2 [11]. After two weeks of selection in 1 mg/ml G418, resistant cells were assayed for the ability to synthesize all 3 G A B A A receptor m R N A s by Northern blotting. The whole cell configuration of the patch clamp technique was used to record the GABA-mediated CIcurrents in A293 cells expressing the cloned G A B A A receptor, cqfl, y 2 or a.fl2, following the experimental proce-
166
dures described elsewhere [3]. Briefly, patch pipettes made of borosilicate glass tubes were fire-polished and showed a resistance of 0.5 2 M.Q when filled with a solution containing (mM) CsC1 140. E G T A I I. MgCle 2, ATP 0.5 and HEPES 10, pH 7,3. The cell-bathing external solution contained (mM) NaCI 135, KC1 5, MgC12 1, CaCL 1.8 and HEPES 5, p H 7.2, (normal saline). G A B A at the concentration of 5/aM in the external solution with or without indicated drugs was applied through a U-tube placed within 100/am of the cells. The current was recorded with an Axopatch 1D amplifier and a CV-4 headstage (Axon Instrument Co.). Bh-1 bath headstage was used to compensate changes in bath potentials. The currents were recorded with a Gould Recorder 220. G A B A currents were measured at the holding potential of - 6 0 mV at room temperature (21 24°C). The dose response curves were analyzed using a logistic equation,
%13272 {LaS']laM 5 ~MGABA
5
I0
50
I00
f. 500 pA
alg2
4o0 8oo" [25opa
ILa~']~M 5 ~MGABA
250
!
I
10
100
200
~ '
0
Ic, = 1..... x [Substrate]"/EC50" + [Substrate]")
(1) 0
where Ic, is the amplitude of the peak C1 current at the given substrate concentration ([Substrate]), Im~xis a maximal response and n is a Hill coefficient. Data fitting was carried out using a computer program (Sigma Plot) of non-linear regression. Data are presented as the mean + S.E.M. from 3 5 separate measurements. Experimental variations in results from different cells were less than 15%. Fig. 1 shows the effect of La -*~ on GABA-mediated CI currents in the cells expressing the o~,flzY2 or 0t~2 subtype of GABAA receptors. In the oq]32T 2 subtype, La > dosedependently potentiated the G A B A (5 /aM) response. Analysis of the dose-response curve using Equation 1 provided the following binding parameters for La>: an EDs0 of 21.3_+3.5/aM, a Hill coefficient of 1.4_+0.4 and the maximal stimulation of 240+16% as normalized to the response by G A B A at 5/aM. In the ot~fle subtype, on the other hand, La > showed no noticeable stimulation of G A B A response until its concentration was raised above 100/aM: Fig. 1 shows the dose-response curve for La > in the a ~ 2 subtype obtained with G A B A at 1/aM, which was similar to that observed with G A B A at 5/aM. The maximal potentiation of the G A B A response by La ~+ reached around 70% at its concentration of 800/aM. This indicates that the 0~2 subtype is lacking a high affinity site for La > associated with the alf1272 subtype. This selectivity of La > is quite opposite to that of Zn 2+ which inhibited the G A B A response in the ~ 6 , with only a marginal action on the ~tfl2Y2 subtype [3, 20]. In this study, we further characterized the potentiation of G A B A currents by La > in the (~1fl272 subtype. Fig. 2 shows the degrees of potentiation by La > (50/aM) as a function of G A B A concentration. The reciprocal plots of
150
0 ~
100
0
50
*P,4
0 1000
[La3+] u Fig. 1. Differential potentiation of GABA-gatedcurrents by La> in the ~qfl2and ottf1272subtypes of GABAAreceptors. The upper panel shows the representative traces for GABA-induced C1 currents in the presence of various concentrations of La> as measured in A293 cells expressing the subtypes, using the whole cell configuration of the patch clamp techniques. The bottom panel shows dose-response curves for La~+potentiation in the presence of 5/aM GABA in ~L//27:(~3)and 1 /aM GABA in ottfl2 (e). The data were fitted to Equation 1and represent the mean and standard errors from 3 cells. [GABA] vs. It, (Fig. 2) show that the stimulatory effect of La 3+ at 50/aM progressively decreased as the concentration of G A B A increased. The analysis of the data with Equation 1 revealed that the major effect of La '~ (50/aM) was to reduce the half maximal concentration for G A B A from 19.3+1 to 12.6+1.1 /aM (n=3) without noticeable effects on the maximal amplitude of the CI- currents and the Hill coefficient (2.6_+0.4 vs. 2.4_+0.4). Although its primary effect was to affect G A B A affinity, La > is not likely to interact directly with G A B A site(s), because G A B A site(s) should be c o m m o n in the both subtypes of GABAA receptors and could not be a contributing factor
167 0.06 IA! s*
GAB^ 2 IIM -
50 IIM 0
250~ GABA
?
Control
• Las+
/
/
ov 0,04
10~M~
.60.,0.,0
o
.~ l
c~
U
GAB^~0.M ~f-
0,02
0.0
0.4
0.8
,-a
0.00 1.2
~ U
Fig. 2. Double-reciprocalplots of GABA-gated currents vs. GABA concentrationswith or without La 3÷. The left panel shows the representative traces for CI currents induced in the presenceof 2, 5, 10 and 20 /.tM GABA with or without 50/.tM La 3÷ as measured in A293 cells expressing%8272subunits. The right panel shows the double reciprocal plots of GABA-induced currents (I,:0 vs. GABA concentrations with (e) or without (o) 50/.tM La3+. for the subtype selectivity. La 3+ may exert its action through an allosteric interaction with a site not shared by the two subtypes. One of the most prominent allosteric site unique to the afl2Y2 subtype is benzodiazepine binding sites [12 14, 18], but it does not appear to be the La 3+ binding site, because excess Ro 15-1788, a classical benzodiazepine antagonist, was unable to block the potentiation of GABA currents by La 3+ (data not shown). Fig. 3 shows typical current-voltage if-V) relationships for the GABA (5/IM)-induced C1- currents with or without La 3÷ at 50 ~tM. La 3+ showed no effect on the reversal potential, 4_+3 mV (n=4), but its potentiation was considerably attenuated at the negative membrane potentials. As reported earlier [21], the afl2Y2 subtype showed an outward rectification in the I - V relationship. The ratio of chord conductances at +40/-40 mV was 2.9+0.4 and decreased to 1.6_+0.3 in the presence of La 3+ at 50/JM. This significant decrease in outward rectification in the presence of La 3- (P
